DESCRIPTION: It is estimated that 7 to 34 percent of pregnant women are abused during pregnancy. Abuse is associated with low birth weight (LBW), which is probably a result of psychosocial stress and unhealthy behaviors, including substance use. Abuse is also connected to increased prenatal hospitalizations and may escalate in the year after birth. The goal of this randomized controlled trial of a nursing case management intervention is to improve maternal and infant outcomes for women who are at high risk for abuse. The specific aims are to: 1) increase women's opportunities to disclose abuse during pregnancy; 2) test the effect of the intervention on reducing psychosocial stress and severity of abuse; 3) decrease the incidence of hospital admissions; 4) estimate the effect of the intervention on reducing LBW; and 5) describe women's transitions through the abuse process during pregnancy and the year after birth and identify what they found helpful. An ethnically diverse sample of 1088 urban and rural women from sites in Oregon and Missouri will be inducted between 16-20 weeks gestation. Data regarding psychosocial stress, abuse, and substance use will be collected at induction (T1) and again between 32-36 week gestation (T2) by research assistants. Women will be randomized to control or treatment groups following the T1 assessment. Women in the intervention group will be categorized as either low risk or high risk for abuse based on their TI data. All low risk women will be given the opportunity to watch the video Faces of Abuse following this initial data collection and provided 24-hour telephone access to a study Nurse Case Manager (NCM), the basic Connections Intervention package. In addition, to the basic package, high-risk women will be offered Connections Plus, which includes the empowerment abuse prevention protocol and individualized case management. The case management will focus on reducing psychosocial stress and the severity of abuse. Medical records will be used to record hospital admissions, the extent of social services received by control group women, and outcome data, including infant birth weight. Study hypotheses will test the differences in psychosocial stress, severity of abuse, number of prenatal hospitalizations and infant birth weights between intervention and control women within and between sites. Data regarding substance use, which will be collected by confidential self-report, will be used to describe and control for this variable in the birth weight analyses. A subset of 20 abused women from each site (N=40), will complete four qualitative interviews during pregnancy and the first postpartum year to describe their transitions through the Abuse process, determine if abuse escalated in the year after delivery, and identify what they found helpful.